DESCRIPTION: Identification of growth regulatory genes that are altered by base mutation, gene amplification, or rearrangement is a major focus of cancer research. For the majority of human solid tumors, including breast cancer, no such oncogene alteration has been found. That such genes exist is strongly suggested by cytogenetic and chromosome hybridization studies. For carcinoma of the breast, loci at 17q22, 20q13, 1q32, 6q21, and 19q seem especially likely to harbor new oncogenes. Previous in vitro studies show that many oncogenes function in one of two ways. Some genes (e.g. RAS) are sufficient to transform immortalized cell lines, whereas others (e.g., C-MYC form foci of tumorigenic, transformed cells only in cooperation with other oncogenes. A large number of genes of the first class have been directly identified by transfection of immortalized NIH3T3 cells with tumor DNA (e.g., RAS, HER-2/neu). However, no assay exists to directly isolate cooperative oncogenes. A major limitation has been the efficiency of transfection for appropriate cell types such as primary rodent cells. That not all genes of the first class can be isolated using NIH3T3 cells is suggested by the behavior of several known oncogenes, including GIP2; BCR-ABL, GL1, and activated alleles of the Wilm's tumor gene, WT-1. These genes fail to transform NIH3T3 cells, but efficiently transform immortalized rat cells, including those immortalized with the adenovirus E1a. This particularly potent oncogene functions by binding several cellular proteins, including the tumor suppressor p105Rb, and cooperates with other genes including RAS, GL1, WT-1, or E1b to give complete transformation of primary rodent cells. In this proposal the investigators describe a method for efficient, stable transfection that uses adenovirus-polylysine conjugates as a transfection reagent. The investigators introduced a plasmid breast cancer cDNA expression library into a well-characterized, E1a- immortalized rat kidney cell line. In contrast to control plasmids, this library induced several large foci of transformed cells. Preliminary characterization of genomic DNA from these clones shows that the known oncogenes RAS or HER-2/neu are not involved. The investigators propose to identify and characterize new E1A-cooperative oncogenes in these cell lines and to test the genes for genetic alteration in breast cancer.